arkadiuszsz / zstd-codec

Zstandard codec for Node.js and Web, powered by Emscripten

zstd-codec

Zstandard codec for Node.js and Web, powered by Emscripten.

Languages

• English

Description

zstd-codec is a binding of Zstandard for Node.js and Browsers, includes JavaScript port of Zstandard compiled with Emscripten.

Installation

npm

npm install zstd-codec

yarn

yarn add zstd-codec

Usage

require module, and instantiate api objects.

const ZstdCodec = require('zsztd-codec').ZstdCodec;
const simple = new ZstdCodec.Simple();
const streaming = new ZstdCodec.Streaming();

• Use Simple API for small data
• Use Streaming API for large data

Simple API

• Using Zstandard's Simple API
  • ZSTD_compress for compress
  • ZSTD_decompress for decompress
• Store whole input/output bytes into Emscripten's heap
  • Available Emscripten's heap size is 16MiB
  • (input.length + output.length) should be less than 12MiB

compress(content_bytes, compression_level)

• content_bytes: data to compress, must be Uint8Array.
• compression_level: (optional) compression level, default value is 3

// prepare data to compress
const data = ...;

// compress
const level = 5;
const compressed = simple.compress(data, level);

// handle compressed data
do_something(compressed);

decompress(compressed_bytes)

• compressed_bytes: data to decompress, must be Uint8Array.

// prepare compressed data
const compressed = ...;

// decompress
const data = simple.decompress(compressed);

// handle decompressed data
do_something(data);

Streaming APIs

• Using Zstandard's Streaming API
  • ZSTD_xxxxCStream APIs for compress
  • ZSTD_xxxxDStream APIs for decompress
• Store partial input/output bytes into Emscripten's heap

const streaming = new ZstdCodec.Streaming();

You can use custom Iterable object on compressChunks / decompressChunks.

compress(content_bytes, compression_level)

• content_bytes: data to compress, must be 'Uint8Array'
• compression_level: (optional) compression level, default value is 3

const compressed = streaming.compress(data); // use default compression_level 3

compressChunks(chunks, size_hint, compression_level)

• chunks: data chunks to compress, must be Iterable of Uint8Array
• size_hint: (optional) size hint to store compressed data (to improve performance)
• compression_level: (optional) compression level, default value is 3

const chunks = [dataPart1, dataPart2, dataPart3, ...];
const size_hint = chunks.map((ar) => ar.length).reduce((p, c) => p + c);
const compressed = streaming.compressChunks(chunks, size_hint); // use default compression_level 3

decompress(compressed_bytes, size_hint)

• compressed_bytes: data to decompress, must be Uint8Array.
• size_hint: (optional) size hint to store decompressed data (to improve performance)

const data = streaming.decompress(data); // can omit size_hint

decompressChunks(chunks, size_hint)

• chunks: data chunks to compress, must be Iterable of Uint8Array
• size_hint: (optional) size hint to store compressed data (to improve performance)

const chunks = [dataPart1, dataPart2, dataPart3, ...];
const size_hint = 2 * 1024 * 1024; // 2MiB
const data = streaming.decompressChunks(chunks, size_hint);

Dictionary API

const zstd_codec = require('zstd-codec');

// compress using trained dictionary
const cdict = new zstd_codec.ZstdCompressionDict(dict_bytes, compression_level);
const compressed = simple.compressUsingDict(data, cdict);

// decompress using trained dictionary
const ddict = new zstd_codec.ZstdDecompressionDict(dict_bytes);
const data = simple.decompressUsingDict(compressed, ddict);

Example

Browser

See the document.

Node.js

TODO: add an example for Node.js.

TODO

• add CI (Travis CI or Circle CI?)
• improve APIs
• write this document
• add how to build zstd with Emsxcripten
• add how to test
• performance test
• add more tests